Background
The COVID-19 pandemic has adversely impacted global health. 1 COVID-19 infections caused widespread mortality, with over 7 million deaths worldwide. 2 Symptoms of COVID-19 infection include fever, cough, shortness of breath, fatigue, headache, pain, sore throat, congestion, nausea, and diarrhea. 3 Physical symptoms may remain significant and impairing following infection, including fatigue, reduced exercise capacity, and decreased ability to perform physical roles.Reference Davis, McCorkell and Vogel 4 , Reference Shanbehzadeh, Tavahomi and Zanjari 5 The pandemic has impacted mental health, characterized by increased psychological distress,Reference Thompson, Stafford and Moltrecht 6 as well as psychiatric symptoms, such as depression, anxiety, and posttraumatic stress disorder.Reference Thompson, Stafford and Moltrecht 6 –Reference Turna, Zhang and Lamberti 9 Downstream effects of the COVID-19 pandemic, including social isolation, financial stressors, health risk, and lifestyle changes, also impacted mental health.Reference Caroppo, Mazza and Sannella 10 –Reference Manchia, Gathier and Yapici-Eser 12
The occupational environment has also changed significantly across multiple industries. Workers experienced significant changes, including working from home, face covering implementation, physical barrier assembly, and improved ventilation.Reference Dehghani, Omidi and Yousefinejad 13 These changes have been associated with a decrease in life satisfaction.Reference Zoch, Bächmann and Vicari 14 Factors associated with occupational environmental stressors include fear of infection or infecting others, job strain, periods of quarantine or isolation, perception of worker rights exploitation, and uncertainty.Reference Giorgi, Lecca and Alessio 15
The mental health impacts of the COVID-19 pandemic have not been uniform across occupational groups. Essential workers, defined as those whose jobs necessitated in-person work, reported significant fear, stress, anxiety, depression, insomnia, burnout, and distress.Reference Beames, Christensen and Werner-Seidler 16 –Reference Sritharan, Jegathesan and Vimaleswaran 18 In addition to being more likely to become infected and die from COVID-19,Reference Gaitens, Condon and Fernandes 19 , Reference Mutambudzi, Niedwiedz and Macdonald 20 essential workers faced other adversities, such as a lack of personal protective equipment, staff shortages, challenges with enforcing regulations, and delays.Reference Nyashanu, Pfende and Ekpenyong 17 , Reference Gaitens, Condon and Fernandes 19 , Reference Gouda, Singh and Gouda 21 , Reference Kavanagh, Pare and Pontus 22 Moreover, when lockdowns were enforced in the USA, testing was largely available to symptomatic or exposed individuals, contributing to stress about asymptomatic spread among essential workers.Reference Thomas, Hamilton and Francis 23 Essential health care workers reported high levels of fear, anxiety, depression, and trauma related to the pandemic, leading to burnout and feeling unable to ensure patient safety.Reference Koontalay, Suksatan and Prabsangob 24 A qualitative study performed with health care workers identified inadequate preparedness, emotional challenges, insufficient equipment and information, and work burnout as related to the pandemic.Reference Koontalay, Suksatan and Prabsangob 24
Educators are a group of essential workers who were significantly impacted throughout the COVID-19 pandemic. Educators are defined as teachers, administrators, paraeducators, and support professionals. Research has demonstrated that teachers reported high rates of anxiety, depression, stress, and burnout related to the COVID-19 pandemic.Reference Agyapong, Obuobi-Donkor and Burback 25 –Reference Spadafora, Reid-Westoby and Janus 28 Many educators were expected to continue teaching, and the struggles of USA educators were outlined in an American Psychological Association (APA) technical report,Reference McMahon, Anderman and Astor 29 which indicated that many educators desired to transfer or quit education, received verbal threats of violence, and experienced physical aggression from students. A related policy brief emphasized that investigating psychological symptoms among educators has never been more timely.Reference McMahon, Anderman and Astor 30
Factors that contribute to the development of psychological distress and disorder have been identified among educators. Lack of collegial support, fear of verbal or physical abuse, significant workload, student behavior, and poor employment conditions were associated with depression, anxiety, and poor mental health.Reference Ferguson, Frost and Hall 31 –Reference Marshall, Smith and Love 33 Among USA educators, increased stressors during the COVID-19 pandemic were related to poorer mental health, and poor mental health was associated with increased difficulty coping and teaching.Reference Baker, Peele and Daniels 34 Additionally, most Wisconsin, USA, educators reported clinically significant depressive and anxiety symptoms.Reference Hirshberg, Davidson and Goldberg 35 Moreover, schoolteachers in Australia reported worsened mental health, citing contributors including increased responsibilities, changes in ways of working, and difficult working conditions.Reference Beames, Christensen and Werner-Seidler 16 In Poland, educators reported at least mild levels of stress, anxiety, and depression, noting contributing factors including blurred professional and personal boundaries and isolation.Reference Jakubowski and Sitko-Dominik 36 The transition from in-person to remote teaching contributed to poorer mental health, with remote teaching associated with higher levels of distress than in-personReference Kush, Badillo-Goicoechea and Musci 37; however, other work has demonstrated high depressive and anxiety symptoms that did not differ based on teaching modality.Reference Carlson, Stegall and Sirotiak 38 Increased workload due to lockdown was suggested as a potential cause of worsened educator mental health.Reference Aperribai, Cortabarria and Aguirre 39 Uncertainty, increased workload, negative perception of the profession, concern for the well-being of others, health struggles, and juggling multiple roles were related to poor mental health among educators in England.Reference Kim, Oxley and Asbury 40
Although quantitative research offers distinct advantages, qualitative and mixed methods research provides nuanced perspectives and context regarding individual experiences of collective events. Specifically, qualitative research methods can clarify the lived experience of individuals.Reference Prosek and Gibson 41 Qualitative research may be particularly helpful in characterizing the lived experience of mental health difficulties.Reference Honey, Boydell and Coniglio 42 COVID-19 and the mental health challenges that it poses may be particularly conducive to qualitative investigation.Reference Tremblay, Castiglione and Audet 43 Even though educators were among the essential workers faced with significant changes and dangers related to COVID-19, qualitative research investigating the effects of the COVID-19 pandemic on educators is limited. Given that many individual and societal experiences secondary to COVID-19 are novel, qualitative investigations are critical to provide contextual nuance. The APA policy briefReference McMahon, Anderman and Astor 30 and technical briefReference McMahon, Anderman and Astor 29 included both qualitative and quantitative data, further emphasizing the importance of both methods to capture educator experiences more fully during the COVID-19 pandemic. In response to the APA reports and the gap in the current literature, this study aims to understand how the COVID-19 pandemic has impacted mental, physical, occupational, and social health among educators.
Method
Participants
Participants (n = 918) were English-reading/speaking adults in the USA who identified as an educator (e.g., teacher, paraeducator, administrator) from kindergarten through college. The average age of the sample was 43 years (SD = 11.3). Most identified as White (96.6%) and female (87.8%). At least one person participated from 46 states of the USA. To be included in the qualitative analyses, a response to one of the 8 qualitative questions was required. There were 145 participants who were excluded due to lack of qualitative response, leaving a sample of 925. Participant responses were randomized into 2 samples for analyses: inductive and deductive. See the demographic characteristics of the two samples in Table 1.
Descriptive characteristics of the inductive(n = 224) and deductive samples (n = 701)

Procedures
Potential participants clicked a link to the survey, which directed them to the survey at qualtrics.com. Congruent with the institutional review board (IRB)-approved protocol, the first page of the survey was a consent letter notifying participants that clicking to the next page indicated consent to participate. The consent specified that participation was voluntary, that the data would be stored on a secure computer drive, and that information provided by participants would not be identifiable. This protocol was approved by the University of Iowa IRB, approval #202007406. Both recruitment and data collection procedures are further detailed alongside quantitative findings.Reference Carlson, Stegall and Sirotiak 38 The qualitative questions that were included are listed below.
-
1. How has your role as an educator during the pandemic impacted your ability to do your job day-to-day?
-
2. How has your role as an educator during the pandemic impacted your professional advancement and career goals?
-
3. How has your role as an educator during the pandemic impacted your mental health?
-
4. How has your role as an educator during the pandemic impacted your physical health?
-
5. How has your role as an educator during the pandemic impacted your social health?
-
6. Do you feel that your school community was receptive and/or responsive to your preferences for educating during the COVID-19 pandemic?
-
7. Were your preferences solicited by a school board, administrative body, county governance, or state governance, and if so, did these preferences appear to be considered in decision-making processes?
-
8. Please feel free to speak to the degree to which you felt your voice as an educator was valued during the pandemic.
Data Analysis
The dataset was divided into an inductive and deductive cohort, using random assignment with a 1:3 scheme (inductive: deductive). In the inductive phase, randomly selected participants were utilized for coding. The purpose of this phase was to develop codes, identify themes, and identify representative data to support codes and themes.Reference Thomas 44 The deductive phase utilized the inductive codes for deductive coding of the data. The purpose was to allow for a structured, systemic analysis to enhance the validity of codes and themes.Reference Fereday and Muir-Cochrane 45 Four coders were involved in analyses. Each coder reviewed all responses to qualitative questions in the inductive dataset. A thematic discussion resulted in the development of relevant codes. Responses to each question were reviewed sequentially to ensure independent consideration for thematic extraction.
The data analytic plan and hypotheses were pre-registered through Open Science Framework (OSF) prior to randomization of responses.Reference Carlson and Thomas 46
Thematic Analysis and Codebook Formation
The themes were formulated based on responses to the 8 qualitative questions, and codes were formed across questions. After identification of themes and the codebook, the inductive dataset was coded by all coders. All codes were reviewed iteratively for accuracy, clarity, and agreement.
Inductive data analyses
In code generation, coders noted themes of interest.Reference Braun and Clarke 47 Individual extractions were collated into themes, ensuring that all data relevant to a theme were included.Reference Braun and Clarke 47 Coders reviewed codes and themes for coherence, accuracy, and refinement.Reference Braun and Clarke 47
Deductive data analyses
Responses were randomly assigned to coders. The deductive codes are reported for interpretation. In addition to random assignment, 30% of each coders’ data were randomly selected for double coding to ensure agreement between coders. All discrepancies were resolved through discussion until agreement was reached.Reference Fereday and Muir-Cochrane 45 , Reference Roberts, Dowell and Nie 48
Results
The results demonstrate that this sample reported declines in physical, mental, and social health, as well as challenges in their professional environment. Educators reported not feeling that their voices were being heard regarding implemented workplace policies. Educators reported taking precautionary measures personally and professionally to prevent infection or spread of COVID-19.
When individuals were recorded as having “not reported” a code, this means that it was not mentioned explicitly. When a participant was recorded as having “reported” a code, this experience was explicitly mentioned and was available for coding. See Table 2 for theme frequencies in each phase.
Frequencies of inductive and deductive codes

Note. Data represent the frequency (n) and percentage (%) of participants in the inductive (n = 224) and deductive (n = 701) samples endorsing each code. Definitions for abbreviated or specific codes are as follows: Alone indicates feeling alone or uncared for; Job change indicates feeling as though the nature of the job has changed (e.g., feeling like a babysitter); Positivity indicates having to stay positive for the kids or behaving incongruent with emotions; Student issues indicates an increase in student behavioral issues; Technology indicates new technology as a factor; and Unsafe policies or spaces indicates policies or physical spaces do not feel safe at work or are not following CDC guidelines.
Themes
COVID-19. Codes observed included: fear of COVID-19 infection, fear of spreading COVID-19, COVID-19 exposure, and COVID-19 infection. Exposure was defined by coders as a known exposure, and infection as a positive test.
Social health. Given the numerous changes in public health recommendations that may have necessitated changes in social behaviors, these themes included responses such as social distancing, lockdowns, isolation, or minimization of contact with others. Codes in this section included limited social interaction, increase or decrease in family time, forgoing activities, virtual activities, and lack of connection or feeling isolated.
Physical health. Derived codes included changes in exercise or physical activity, weight, and/or diet changes, sleep changes, somatic symptoms, fatigue, and lack of access to health behavior facilitators (e.g., gyms or health care visits). Because of the regulations that varied by state, some of these activities were forbidden or strongly recommended against. Others reported forgoing activities to minimize risk.
Mental health. Derived codes included psychological experiences, including depression, anxiety, suicidality, and helplessness. This section highlights changes to existing psychiatric medications or starting new psychiatric medication. Themes of seeking new psychotherapeutic services or desire but inability to seek therapy were observed. This section included stress, grief, uncertainty, self-reflection, and appreciation of solitude.
Professional well-being. Derived codes included a sense that the education being provided was inadequate. A theme of having unsafe policies or physical spaces, such as classrooms being unfit for social distancing guidelines, was observed. Themes regarding student issues, parental disrespect, financial concerns, and either the presence or lack of administrative support were observed. Educators reported feeling overworked and having increased responsibilities. Educators reported insufficient personal protective equipment (PPE) to safely work in-person. Themes of career dissatisfaction and plans for career change were present. Some reported having shifted careers. On the other hand, some reported feeling validated in their career choice and had an increased sense of belonging. Another theme was having one’s voice heard by administrators; some reported having a role in decision-making processes, whereas others voiced opinions and were ignored. Others were not asked for feedback and were informed of plans.
Precautions taken. Derived codes highlighted the precautions that educators were taking to avoid becoming infected with or spreading COVID-19. Themes included disinfecting surfaces, social distancing, and altering eating and shopping behaviors.
Deductive Findings
Participants reported fear of spreading (13.1%) or becoming infected with (11.6%) COVID-19, with some reporting a known exposure or infection at work (0.6%). When connected with CDC county-level data, most participants who also completed the quantitative data portion of this study (88.0%) reported living in an area with a “high” rate of COVID-19 transmission.Reference Carlson, Stegall and Sirotiak 38 Most reported limiting social interactions (72.0%), and 27.8% reported feeling lack of connection or isolation. Moreover, 49.5% chose not to attend certain activities, and 29.8% reported virtual engagement in activities. Changes in mental and physical health were also reported. New-onset symptoms were reported, with 11.9% of participants reporting new-onset somatic symptoms and 23.9% reporting new-onset mental health symptoms. Less often noted were exacerbations of physical illness (3.1%) or mental health difficulties (2.8%). Participants reported decreased physical activity (16.4%), weight gain (12.4%), and negative changes in diet (9.7%). Others reported positive health behavior changes (e.g., positive changes in diet [1.6%], increased self-care [2.0%], and increased exercise or physical activity [4.9%]).
Alongside mental health difficulties, many specific emotions and symptoms were reported, including anxiety (20.9%), stress (38.0%), frustration (12.9%), and uncertainty (12.9%). Many reported feeling helpless or hopeless (18.5%) and alone (21.6%). Several personal protective behaviors were reported, including altering eating behavior (44.4%) and personal PPE use (19.5%). Many reported a lack of administrative support across district-level (20.1%), building-level (4.0%), and unspecified levels of administrative support (15.4%). Others reported administrative support (district-level: 4.1%; building-level: 5.5%). Many reported their concerns were not heard when voiced (28.2%), while others reported feeling heard (9.9%). Others were told of plans, not asked (10.0%). Some reported being told of plans, not asked for their concerns, and their voice was not heard (3.5%).
Underscoring the competing demands that educators faced, 34% reported increased work responsibility, and 21% reported that the nature of their job changed. Some participants felt overworked (17.5%), and others reported that expectations had increased in a way that was unachievable (16.0%). Some participants reported feeling that they were delivering an inadequate education, and almost one-quarter reported unsafe policies or spaces at work (23.0%). Some participants reported work dissatisfaction (13.7%), and participants reported feeling work-related resentment (27.0%).
Discussion
The purpose of this study was to understand how the COVID-19 pandemic impacted domains of health (mental, physical, and social), day-to-day functioning, and career satisfaction among educators in the USA. Although research detailed the physical and psychological health impacts of COVID-19 on educators worldwide, limited work has focused on educators’ lived experiences. The present study sought to derive themes inductively and apply the derived codes to the deductive dataset. The authors describe educator experiences across domains of health as detailed by educators in November and December of 2020.
The findings demonstrated deleterious mental and physical health impacts, where participants reported feeling demoralized and distressed. Participants endorsed occupational changes, including planning or desiring to leave the profession, in part due to new responsibilities and insufficient time. Educators reported increased expenses for work supplies, such as cleaning supplies and PPE. Many educators reported isolation in service of protecting others, with some living separately from family or not seeing relatives, limiting access to supports. See Table 3 for deidentified quotes sampled from educators in the deductive phase of analyses.
Deidentified quotes from participants in the deductive phase, grouped by themes

Note. Quotes that include [bracketed text] were altered to remove any potentially identifying information.
Our results align with literature noting worsened physical and mental health during the COVID-19 pandemic among educators worldwide.Reference Thompson, Stafford and Moltrecht 6 –Reference Turna, Zhang and Lamberti 9 , Reference Zhao, Li and Sang 49 Decreased physical activity, increased sedentary behavior, weight gain, negative dietary changes, and sleeping less were noted among our sample, fitting with literature noting a decrease in healthy lifestyle behaviors during the COVID-19 pandemic.Reference Zhao, Li and Sang 49 , Reference Park, Zhong and Yang 50 Somatic symptom burden was also significant, fitting with literature noting increased somatic symptom burden during the COVID-19 pandemic.Reference Engelmann, Löwe and Brehm 51 Anxiety, stress, depression, uncertainty, and feeling hopeless or helpless were reported in our sample, aligning with extant research.Reference Thompson, Stafford and Moltrecht 6 –Reference Turna, Zhang and Lamberti 9 , Reference Zhao, Li and Sang 49 Some participants noted increased self-reflection, fitting with research observing improved self-concept, autonomy, and psychological well-being among some during COVID-19 lockdowns.Reference Paterson and Park 52 As with a significant proportion of the world’s population,Reference Kim and Jung 11 our sample reported limiting social interaction and forgoing activities. Although concerns about developing or spreading COVID-19 were frequently noted among our sample, confirmed exposure and infection were relatively low, though infection was not directly assessed herein.
Past literature has noted educators as essential workers affected by the pandemic with reports of lack of colleague support, increased workload, and poor employment conditions.Reference Agyapong, Obuobi-Donkor and Burback 25 , Reference McMahon, Anderman and Astor 29 , Reference Ferguson, Frost and Hall 31 , Reference Kovess-Masféty, Rios-Seidel and Sevilla-Dedieu 32 , Reference Klusmann, Aldrup and Roloff-Bruchmann 53 –Reference Stang-Rabrig, Brüggemann and Lorenz 55 Our results further illustrate the lived experience of educators during this time. The participants noted increased responsibility, with failure to meet professional self-expectations, perception of providing inadequate education, being overworked, and feeling that the nature of the job had changed. These reports mirror wider literature noting increased responsibilities, juggling multiple roles, and the blurring of professional-personal boundaries.Reference Beames, Christensen and Werner-Seidler 16 , Reference Jakubowski and Sitko-Dominik 36 , Reference Aperribai, Cortabarria and Aguirre 39 , Reference Kim, Oxley and Asbury 40 , Reference Stang-Rabrig, Brüggemann and Lorenz 55 The APA technical report highlighted concerns regarding the safety of educators regarding physical violence, verbal abuse, and risk of COVID-19 infection.Reference McMahon, Anderman and Astor 29 The present findings describe unsafe policies and physical spaces at work related to the infectious nature of COVID-19, further highlighting educator distress.
Educators noted varied experiences regarding support in the workplace. Building-level administrative support was reported more frequently than district-level administrative support, indicating that those in closer proximity to our sample (primarily teachers) provided more support. Several participants voiced concerns to superiors about the occupational environment with varying levels of success. The highest proportion who voiced concerns did not feel those concerns were heard. The perception of difficult working conditions has been associated with mental health difficulties among educators,Reference Beames, Christensen and Werner-Seidler 16 , Reference Ferguson, Frost and Hall 31 , Reference Kovess-Masféty, Rios-Seidel and Sevilla-Dedieu 32 , Reference Heikonen, Ahtiainen and Hotulainen 56 and occupation-related concerns and lack of administrative support noted among some in our sample mirror this finding. Perhaps associated with these issues, resentment of employers was observed. Some educators noted desire or plan to shift away from educational careers, whereas others intended to pursue early retirement. These findings were also demonstrated by the APA, with 49% of US educators surveyed having a desire to leave the profession.Reference McMahon, Anderman and Astor 29
Finally, pandemic precautions were reported during early stages of the COVID-19 pandemic,Reference Girum, Lentiro and Geremew 57 and our sample also endorsed protective behaviors. The most common protective behaviors were increasing ventilation, disinfecting surfaces, social distancing, using PPE, and altering shopping behaviors, fitting with findings in the wider population from this time.Reference Masters, Shih and Bukoff 58 –Reference Truong and Truong 60 However, other precautions seemed specific to the occupational environment, such as monitoring masking among students and altering eating behaviors (e.g., not eating in classroom or break room).
Limitations
There are several limitations to this research. The sample was primarily White, non-Hispanic females. Even though this aligns with the population of USA educators, our sample lacks diversity in race, ethnicity, and gender.Reference McMahon, Anderman and Astor 29 , 61 The study advertisement was posted on Facebook and Reddit. This may have resulted in a lack of responses from educators who did not use social media. The study was cross-sectional; therefore, no longitudinal conclusions can be drawn. The data did not capture duration or intensity of experiences. Data were collected via survey to maximize participant ease, but no follow-up or clarifying questions were asked.
As our study involved educator responses during 2020, our findings do not apply to educators’ experiences during other phases of the pandemic. Longitudinal work could be utilized to assess the experience of educators over time. While our work analyzed written qualitative responses, other qualitative interviews or focus groups may allow follow-up and in-depth understanding of the educator experience. Future work could assess long-term implications of the COVID-19 pandemic on educators.
Implications
Given the expressed mental health concerns, systemic emphasis on utilization of mental health care resources may be a helpful future direction. Providing resources in the workplace would improve access. One study in Colorado demonstrated workplace integration, with eight schools providing social-emotional competencies workshops in the workplace, and demonstrated benefit.Reference Fitzgerald, Shipman and Pauletic 62
Beyond individual-level interventions, systems-level interventions should also be considered. School psychologists are often confused as resources for the students and teachers, which is typically not the case. School psychologists serve as an invaluable resource for students struggling with issues related to learning and school, but oftentimes there are few (if any) resources for educator mental health within the school system.Reference Lever, Mathis and Mayworm 63 However, when providing a mental health care specialist for educators has been tested, mindfulness has improved, and stress has decreased.Reference Molina, Lemberger-Truelove and Zieher 64 Although this research is promising, it does not address the needs of educators in areas that are not well-resourced. Districts that cannot provide an on-site mental health care specialist for educators should consider providing low-cost, virtual mental health care for employees.
The “Supporting the Mental Health of Educators and Staff Act of 2023” highlights the incorporation of mental health care services in the workplace and proposes substantial funding to support this type of programming at a policy level within the USA. 65 Including mental health care screenings in employee health screenings would further the goal of normalizing checking in about mental health and well-being.Reference Krist, Phillips and Sabo 66 Having referral options following screening is of critical importance.
At the systemic level, lack of agency or voice was described by educators in our sample. Many wrote about decisions being made on their behalf at the building and district levels. Encouraging representation of educators as key stakeholders in these choices would increase agency. Increasing spending for educator health would allow for identification of predictors of workplace satisfaction among educators.Reference Sorensen, Dennerlein and Peters 67 , Reference Watlington, Shockley and Guglielmino 68 This would elucidate problem areas and improve teacher retention, leading to fiscal benefits and downstream improved student achievement.Reference Sorensen, Dennerlein and Peters 67 , Reference Watlington, Shockley and Guglielmino 68
Conclusion
In summary, the present study examined USA educators’ experiences. Many of the themes aligned with issues that have been highlighted in media and policy beyond the COVID-19 pandemic, including lack of perceived support by administration and inadequate time allotted for responsibilities. Our findings demonstrated negative impacts on educators’ physical and psychological health, as well as a desire to feel heard in decision-making. Further intervention research is needed to address educator mental and physical health at different levels of intervention (individuals, systems, policies). As educators shape the future of students and communities worldwide, it is imperative that their well-being is a top priority.
Data accessibility
Deidentified or aggregated data will be made available upon reasonable request, congruent with the approved IRB application.
Acknowledgments
Many thanks to Tyler Kuhn and the THRIVE Lab for their assistance in this study. We acknowledge the many educators who took valuable minutes of their limited time to complete the survey and share their experiences with us. Finally, we would like to acknowledge Hannah for inspiring the study’s conceptualization and providing feedback on the survey—your resilience is remarkable.
Author contribution
Authors ASC and EBKT designed the study and collected the data. Authors ASC, ZS, and SO completed analyses. Authors ASC, ZS, and SO prepared the first draft of the manuscript with contributions from EBKT. All authors revised and approved the final manuscript.
Funding statement
This research was supported in part by the National Institute of Health T32 pre-doctoral training grant: T32GM149386 (Alyssa Schneider Carlson). NIH did not have any role in the study design, collection, analysis, or interpretation of the data, writing of the manuscript, or the decision to submit the paper for publication.
Competing interests
All authors declare that they have no conflicts of interest.